#Part of settings that should not be subject to changes.

#Source the fixed part that will be in the environment (coming from BLPdata)====
source("CF_MMX_settings_fixed_env.R") 


#
#MCES====
#
fixed.settings.mces <- data.frame(a2=c(0,0,-1,-1,0),  
                         mu.b=rep(1,5), # 
                         b2=c(0,0,0,0,1),  # used only to control old setting 5
                         b.adj=c(1,1,1,NA,NA),  # adjust the mean b for the x in demand
                         sigma.b.adj=c(0,1,1,NA,NA),  # adjust the sigma b for the x in demand
                         mu.c=rep(0,5),  # there is a constant in the cost function in BLP
                         sigma.xi=c(SDxi$sd.xi,NA,NA),  #taken from our backed out xi in BLP data
                         gamma1=c(1,1,1,NA,NA), # determines how observable attributes affect marginal costs, =1 in BLP
                         gamma2=c(0.2,0.2,0.2,NA,NA), # >0 --> xi does enter costs, xi is correlated with RHS, downward bias in eta
                         income.slope=rep(0,5),
                         gini.slope=rep(0,5),
                         quality.slope=rep(0,5))

#
#MLOG====
#
fixed.settings.mlog <- data.frame(a2=c(0,0,-1,-1,0),  
                                  mu.b=rep(1,5), # 
                                  b2=c(0,0,0,0,1),
                                  b.adj=c(1,1,1,NA,NA),
                                  sigma.b.adj=c(0,1,1,NA,NA),  # adjust the mean b for the x in demand
                                  mu.c=rep(0,5),
                                  sigma.xi=c(SDxi$sd.xi,NA,NA),
                                  gamma1=c(1,1,1,NA,NA), # determines how observable attributes affect marginal costs, =1 in BLP
                                  gamma2=c(0.2,0.2,0.2,NA,NA), # >0 --> xi does enter costs, xi is correlated with RHS, downward bias in eta
                                  income.slope=rep(0,5),
                                  gini.slope=rep(0,5),
                                  quality.slope=rep(0,5))








